Competitive Promoter-Associated Matrix Attachment Region Binding of the Arid3a and Cux1 Transcription Factors

diseases

Article Competitive Promoter-Associated Matrix Attachment Region Binding of the Arid3a and Cux1 Transcription Factors

Dongkoom Kim 1,2, Christian Schmidt 1,3, Mark A. Brown 4 and Haley Tucker 1,* 1 Molecular Biosciences, Institute for Cellular and Molecular Biology, University of Texas at Austin, Austin, TX 78715, USA; [email protected] (D.K.); [email protected] (C.S.) 2 Atreca, Inc., Redwood City, CA 94063, USA 3 Department of Biomaterials and Healthcare, Division of Life Science and Bioprocesses, Fraunhofer-Institute for Applied Polymer Research (IAP), 14476 Potsdam-Golm, Germany 4 Department of Clinical Sciences, Colorado State University, Fort Collins, CO 80523, USA; [email protected] * Correspondence: [email protected]; Tel.: +1-512-475-7705

Received: 15 November 2017; Accepted: 5 December 2017; Published: 10 December 2017

Abstract: Arid3a/Bright/Dril1 is a B cell-speciﬁc transactivator that regulates immunoglobulin heavy chain (IgH) gene transcription by binding promoter and enhancer-associated matrix attachment regions (MARs) within the IgH gene locus. Promoter MAR-mediated Arid3a transactivation is antagonized by direct competition of MAR binding by Cux1/CDP—a ubiquitously expressed repressor originally termed NF-µNR. We report that the NF-µNR complex includes Arid3a in B cells but not in non-B cells through mobility shift assays. The binding activity of NF-µNR and Arid3a in B cells is reciprocally altered during the cell division cycle and by the B cell mitogen lipopolysaccharide LPS. LPS treatment had no effect on Arid3a localization but increased its total abundance within the nucleus and cytoplasm. We show that this increased level of Arid3a is capable of displacing Cux from the MARs to facilitate IgH gene transcription. Finally, we showed that the MARs (termed Bf150 and Tx125) associated with the VH1 rearranged variable region expressed in the S107 murine plasmacytoma, can repress reporter gene transcription in non-B cells and that they can relieve the repression mediated by Eµ enhancer in B cells. These results have signiﬁcant implications for early human development and demonstrate that MARs in IgH locus, NF-µNR and Arid3a regulate IgH gene expression in a concerted fashion. This paves the way for future studies examining the misregulation of this pathway in pediatric disease.

Keywords: immunoglobulin heavy chain; Arid3a; NF-µNR; matrix-attachment region (MAR); transactivation

1. Introduction Immunoglobulin heavy chain (IgH) gene transcription has served as a model system to study tissue-specific and developmental stage-specific gene regulation. One of the prominent features of IgH gene regulation is that numerous regulatory elements are scattered across the expansive IgH locus (Figure1; Figure S1). These include the V H promoter, the intronic enhancer (Eµ) and the 3’ enhancers [1–4]. These elements are composed of various transcription factor binding motifs which, in some cases, are flanked or are proximal to matrix attachment regions (MARs) [5,6]. Considerable effort has focused on elucidating the functions of each of these elements. For example, in some contexts, either the promoter or Eµ, alone, is sufficient for tissue-specific expression of IgH. Yet the combination of the two elements achieved by somatic VDJ recombination leads to the high levels of transcription required to achieve antibody levels necessary for defense [6,7]. In addition, the contribution of critical

Diseases 2017, 5, 34; doi:10.3390/diseases5040034 www.mdpi.com/journal/diseases Diseases 2017, 5, 3410.3390/diseases5040034 2 of 16 Diseases 2017, 5, 34 2 of 16 transcription required to achieve antibody levels necessary for defense [6,7]. In addition, the motifscontribution within of each critical of motifs these elements, within each such of thes as thee elements, octamer such within as the the octamer promoter within and Etheµ, promoter has been studiedand Eμ, extensivelyhas been studied [8]. extensively [8].

Figure 1. Mouse IgH locus and thethe S107S107 VVHH1 promoter region. ( A)) IgH gene locus consists of gene segment clusters in the germ line. Boxes indicate ex exonsons and gray ovals are enha enhancers.ncers. Arrows indicate germ-line transcription; (B) Gene rearrangement assembles a functionalfunctional VDJ exon and juxtaposes the

VH-associated-associated promoter promoter with with the the intronic intronic enhancer, enhancer, E Eμµ; ;((CC) )The The promoter promoter of of the the VH V1H variable1 variable gene gene of ofthe the S107 S107 family family is iscomposed composed of of a acore core promoter promoter and and two two MARs. MARs. Closed Closed triangles triangles and open boxes represent cis protein binding sites within MARs or core promoter, respectively. Transcription factors are shown under the motifs toto whichwhich theythey bind.bind.

In spitespite of of intense intense interest interest in thein VtheH coreVH promoter,core promoter, far less far is knownless is aboutknown the about promoter-associated the promoter- MARs.associated The MARs. MARs The (operationally MARs (operationally termed Bf150 termed and Tx125 Bf150 and and residing Tx125 and ~150 residing and ~125 ~150 bp upstreamand ~125bp of theupstream VH1 promoter) of the VH were1 promoter) reported were to be reported required to for be maximal required transcriptional for maximal transcriptional induction of the induction murine S107of the V Hmurinelocus byS107 mitogens VH locus and by certain mitogens cytokines and certain [9]. These cytoki MARsnes [9]. contain These specific MARs bindingcontain sitesspecific for Arid3abinding and sites its for antagonist, Arid3a and CDP/Cux its antagonist, [10,11 CDP/Cux]. Since Arid3a [10,11]. is Since expressed Arid3a primarily is expressed in B primarily cells, these in dataB cells, raised these the data possibility raised the that possibility Bf150 and that Tx125 Bf150 might and function Tx125 might in a lineage-specific function in a lineage-specific manner. Stable transfectionmanner. Stable of atransfection reporter containing of a reporter the VcontainingH1 core along the V withH1 core Bf150 along and with Tx125 Bf150 into and a plasma Tx125 into B cell a lineplasma (J558) B cell achieved line (J558) significant achieved levels significant of transcription levels of transcription that could be that further could stimulated be further by stimulated Arid3a [10 by]. However,Arid3a [10]. Arid3a However, could Arid3a not transactivate could not transactivate the reporter the through reporter a concatamer through a ofconcatamer Bf150. These of Bf150. and related These experimentsand related experiments led to the speculation led to the speculation that the appropriate that the appropriate arrangement arrangemen or “germline”t or “germline” contextof context these elementsof these elements is required is required for Arid3a for to Arid3a properly to properly engage and engage activate and them.activate However, them. However, a transgenic a transgenic study of S107study transcription of S107 transcription argued that argued Bf150 that and/or Bf150 Tx125 and/or were Tx125 dispensable were dispensable for VH1 activationfor VH1 activation and that and the corethat the promoter core promoter region was region sufficient was sufficient to confer to lymphocyte-specific confer lymphocyte-specific expression expression [12]. Therefore, [12]. Therefore, the roles ofthe the roles Bf150 of the and Bf150 Tx125 and MARs Tx125 in MARs VH1-mediated in VH1-mediated transcription transcription remained remained unclear. unclear. Eµμ,, residingresiding inin thethe intronintron betweenbetween VDJ and Cµμ,, isis composedcomposed ofof corecore enhancerenhancer andand thethe flankingflanking MARs [6[6].]. TheThe E µEμis knownis known to be to sufficient be sufficient for lymphoid-specific for lymphoid-specific expression expression of IgH gene of IgH [13]. gene However, [13]. AguileraHowever, et Aguilera al. showed et al. that showed some Bthat cell some lines B maintain cell lines high maintain expression high expression of endogenous of endogenous IgH locuswith IgH locus with Eμ deletion, implying the existence of other functionally redundant activating elements in

Diseases 2017, 5, 34 3 of 16

Eµ deletion, implying the existence of other functionally redundant activating elements in the IgH locus [14]. Meanwhile, Kaplan et al. demonstrated that Arid3a interacts Eµ-flanking MARs and transactivates flanking genes [10]. EMSA experiments demonstrated that Bf150 and Tx125 produce two major DNA-protein complexes in B cells: An Arid3a-containing complex and a NF-µNR-containing complex. It was proposed [9,15] that Arid3a and NF-µNR compete for binding to common “P sites” within each of these MARs, with the balance leading to activation or repression, respectively. Consistent with this model, CDP/Cux was identified as the functional component of NF-µNR, and competition with CDP/Cux abrogated Arid3a-DNA binding [15]. In several cases, CDP/Cux was documented to function by competing with activators for DNA binding sites [9,16]. In other cases, CDP/Cux was implicated to actively repress transcription by recruiting HDAC1 [17]. The activity of CDP/Cux is regulated during the cell cycle, as specific DNA binding was restricted to the G1/S transition and S phase [18]. However, in several cases, CDP/Cux DNA binding activity was shown to disappear as cells become fully differentiated [19]. An added complexity that was not considered in previous studies is the contribution of the related ARID transcription factor, Arid3b/Bdp. Although poorly characterized, the single study published on this Arid3a paralog demonstrated that in vitro translated Arid3b could also bind to bf150 and Tx125 [20]. We showed previously that Arid3b interacts with Arid3a to modulate its localization [21,22]. Yet it remained unclear as to whether Arid3a-Arid3b heteromers bind to the promoter-associated MARs, or if their homomeric and Arid3a heteromeric complexes serve redundant or specific functions. In this study, we have investigated features and components of the Arid3a and NF-µNR complexes through EMSA/super-shift assays in order to better understand the function of promoter-associated MARs in IgH transcription. We demonstrate that these complexes are sensitive to cell cycle progression and to mitogen stimulation. We also employed a non-B cell transcription assay to examine the effect of Arid3a on the promoter- and enhancer-associated MARs in the absence of other B cell specific regulators.

2. Materials and Methods All experiments for this study were performed in triplicate.

2.1. Cell Culture

M12.4, a mouse lymphoma [23], and BCL1, mouse leukemia [24] cells were cultured in RPMI medium supplemented with 10% fetal calf serum (FCS). The adherent cell lines, Cos-7 and NIH 3T3, were grown in Dulbecco modiﬁed Eagle medium (DMEM) supplemented with 10% FCS under standard conditions. To examine the starvation effect, BCL1 cells were grown in RPMI complete medium for 3 days with or without the daily supplement of fresh medium. To stimulate BCL1 cells with LPS, the cells were grown in RPMI complete medium supplemented with 10 µg/mL of LPS for 2–3 days.

2.2. Transfections and Retroviral Transduction NIH3T3-Arid3a stable lines were established using the Phoenix retroviral system according to published protocols (https://web.stanford.edu/group/nolan/_OldWebsite/retroviral_systems/phx. html) as previously described [21]. Luciferase-expressing NIH 3T3 stable cell lines (NIH 3T3 50 MAR-Luc and NIH 3T3 50 MAR-Luc-Eµ) were established by co-transfecting NIH 3T3 cells with a neomycin-resistant gene-bearing mini vector and the pGL3-50 MAR construct using FuGene6 transfection reagent (Roche Diagnostics). Neomycin-resistant luciferase-positive clones were selected with 600 µg/mL of G418 for 12 days. The 50 MAR luciferase-expressing NIH 3T3 stable cell lines were then transiently transfected with Arid3a constructs using FuGene6 transfection reagent. Brieﬂy, 1.5 × 105 to 3 × 105 cells were Diseases 2017, 5, 34 4 of 16 cultured overnight in six-well plates containing 2 mL of medium. The cells were transfected with 0.3–1 µg of plasmid DNA and cultured for 24–48 h.

2.3. Nuclear Protein Extraction Cytoplasmic and nuclear extracts were prepared according to the method of Johnson et al. [25]. B cells (~1 × 107) were collected, washed twice with PBS, and pellets were suspended in 100 µL of sucrose buffer I-A (100 mM Tris-Cl, pH 8.0; 0.32 M sucrose; 3 mM CaCl2; 2 mM Magnesium acetate; 0.1 mM EDTA; 1 mM DTT; 0.5 mM PMSF and 0.5% NP-40) supplemented with 0.1 µL of protease inhibitor cocktail (Sigma, St. Louis, MO, USA). Nuclei were separated from the soluble cytoplasmic fraction by centrifugation at 2000 rpm for 2 min (Fisher Scientiﬁc, Model 59-A, Hampton, NH, USA). Nuclei were suspended in 20 µL of low salt buffer (20 mM HEPES, pH 7.9; 25% glycerol; 1.5 mM MgCl2; 20 mM KCl; 0.2 mM EDTA; 0.5 mM DTT and 0.5 mM PMSF), and then 20 µL of high salt buffer (20 mM HEPES, pH 7.9; 25% glycerol; 1.5 mM MgCl2; 800 mM KCl; 0.2 mM EDTA; 1% NP-40; 0.5 mM DTT and 0.5 mM PMSF) were added dropwise. Then, the extract was diluted with 100 µL of diluent (25 mM HEPES, pH 7.6; 25% glycerol; 0.1 mM EDTA and 0.5 mM PMSF). The mixture was centrifuged at 12,000 rpm for 15 min at 4 ◦C (Eppendorf, 5417R Hamburg, Germany), and the supernatant was used for EMSA.

2.4. Electrophoretic Mobility Shift Assays (EMSA) In vitro DNA binding reactions were conducted in a total volume of 25 µL of binding buffer (10 mM HEPES, pH7.9; 10% glycerol (v/v); 50 mM NaCl; 0.5 mM EDTA; 0.02% Tween-20; 80 ng/µL poly-(dI/dC) and 0.25 mM PMSF), containing 2 µg of nuclear extract and ~80,000 cpm of Arid3a-speciﬁc probe: 5’-end labeled, gel-puriﬁed fragments spanning either of the VH S107 promoter-associated MARs (Bf150 or Tx125) as previously described [26]. Samples were incubated for 20 min at room temperature and then resolved on 4% polyacrylamide gels. For the super-shift assays, indicated antibodies or antisera were added to the above binding reaction following probe-extract incubation. Samples were placed on ice for 30 min before loading onto gels.

2.5. Cell Cycle Synchronization and Cell Cycle Analysis

To synchronize the cell cycle at G1/S phase, BCL1 cells were grown in RPMI medium supplemented with 0.5% FCS for 18 h and then shifted to RPMI medium supplemented with 10% FCS and 1 mg/mL of aphidicolin (Sigma) for 16 h. The cells were washed with phosphate-buffered saline (PBS) to remove the cell cycle inhibitor and grown in RPMI medium supplemented with 10% FCS. B cells were harvested every 2 h, ﬁxed in ethanol, washed with PBS, incubated for 30 min with 10 µL of 50 µg/mL propidium iodide (Sigma) and 10 µL of 10 mg/mL RNase A, and then analyzed by ﬂow cytometry for DNA content.

2.6. In Vitro Transcription/Translation Arid3a was expressed using the TNT Coupled Transcription/Translation Systems (Promega, Madison, WI, USA) following the manufacturer’s instructions. Protein integrity and level was assessed by anti-Arid3a Western blotting.

2.7. Luciferase Assays NIH 3T3 50 MAR-Luc cells were transfected with pBabe-Arid3a wild type. Twelve or 24 h later, the luciferase activity was measured within 10 µg of whole-cell lysate using the luciferase reporter assay system (Promega) according to the manufacturer’s instructions. To evaluate the effect of Arid3a on luciferase reporter gene activity, we calculated the relative luciferase activities of Arid3a-expressing cells compared with that of NIH 3T3 50 MAR-Luc. Diseases 2017, 5, 3410.3390/diseases5040034 5 of 16

3. Results

3.1. Components of the Arid3a and NF-μNR Complexes in B Cells In previous studies, two major DNA binding complexes in B cells have been detected on the two MARs, Tx125 and Bf150, upstream of the IgH promoter [9,27]. NF-μNR forms a slowly migrating complex, and its functional component, CDP/Cux, is associated with IgH repression [17]. Arid3a forms a more rapidly migrating complex which is associated with locus activation. None of the previousDiseases 2017 studies,, 5, 34 however, have directly compared the relative binding of Tx125 and bf150, nor 5have of 16 they considered the potential contribution of Arid3b. 3. ResultsTo re-examine these issues, we prepared nuclear extracts from the mature B cell lines, BCL1 and M12.4, and performed EMSA. As shown in Figure 2, the Arid3a-containing complex has a relatively higher3.1. Components affinity for of Bf150 the Arid3a than andfor NF-Tx125.µNR Conversely, Complexes inthe B NF- CellsμNR complex exhibits higher affinity for Tx125 than for Bf150. The differential affinity of Arid3a and NF-μNR were observed in both M12.4 In previous studies, two major DNA binding complexes in B cells have been detected on the two and BCL1 cells (Figure 2). In addition, we consistently observed that anti-Cux antiserum slightly MARs, Tx125 and Bf150, upstream of the IgH promoter [9,27]. NF-µNR forms a slowly migrating interfered with Arid3a-DNA binding in BCL1 but not in M12.4 (Figures 2 and 3). This suggested that complex, and its functional component, CDP/Cux, is associated with IgH repression [17]. Arid3a the complexes in the two B cell lines are not identical. forms a more rapidly migrating complex which is associated with locus activation. None of the Next, we examined the components of each complex by performing super-shift assays with previous studies, however, have directly compared the relative binding of Tx125 and bf150, nor have various antibodies. The anti-Arid3a antisera employed has been shown previously [15] to possess no they considered the potential contribution of Arid3b. cross-reactivity against CDP/Cux, Arid3b, or any other protein tested. As expected, anti-Arid3a To re-examine these issues, we prepared nuclear extracts from the mature B cell lines, BCL1 and treatment completely ablated the Arid3a complex in BCL1, shifting it to the well of the gel (Figure 3). M12.4, and performed EMSA. As shown in Figure2, the Arid3a-containing complex has a relatively In addition, and unanticipated from previous reports, anti-Arid3a consistently super-shifted the NF- higher affinity for Bf150 than for Tx125. Conversely, the NF-µNR complex exhibits higher affinity for μNR complex formed on either probe to a slightly slower mobility (Figures 2 and 3). This subtle NF- Tx125 than for Bf150. The differential affinity of Arid3a and NF-µNR were observed in both M12.4 μNR super-shift with anti-Arid3a was also observed in M12.4 cells (Figure 2). The super- and BCL1 cells (Figure2). In addition, we consistently observed that anti-Cux antiserum slightly shift/ablation of the NF-μNR complex with anti-Cux confirmed that the super-shifted complex with interfered with Arid3a-DNA binding in BCL1 but not in M12.4 (Figures2 and3). This suggested that anti-Arid3a is the NF-μNR complex. the complexes in the two B cell lines are not identical.

Figure 2. NF-µNR and Arid3a bind to V 1 promoter-associated MARs Tx125 and Bf150 with different Figure 2. NF-μNR and Arid3a bind to VHH1 promoter-associated MARs Tx125 and Bf150 with different afﬁnities. BCL1 (A) and M12.4 (B) nuclear extracts were prepared and subjected to EMSA using32 affinities. BCL1 (A) and M12.4 (B) nuclear extracts were prepared and subjected to EMSA using P- 32P-labeled Bf150 or Tx125 fragments as probes. Super-shift assays were performed with antisera or labeled Bf150 or Tx125 fragments as probes. Super-shift assays were performed with antisera or mAB mAB (α) listed above the corresponding lanes. Plane arrows indicate the mobility of the NF-µNR (α) listed above the corresponding lanes. Plane arrows indicate the mobility of the NF-μNR complex; complex; arrows with balls, the Arid3a complex. No NE: free probe, PI: pre-immune serum. arrows with balls, the Arid3a complex. No NE: free probe, PI: pre-immune serum.

Next, we examined the components of each complex by performing super-shift assays with various antibodies. The anti-Arid3a antisera employed has been shown previously [15] to possess no cross-reactivity against CDP/Cux, Arid3b, or any other protein tested. As expected, anti-Arid3a treatment completely ablated the Arid3a complex in BCL1, shifting it to the well of the gel (Figure3). In addition, and unanticipated from previous reports, anti-Arid3a consistently super-shifted the NF-µNR complex formed on either probe to a slightly slower mobility (Figures2 and3). This subtle NF-µNR super-shift with anti-Arid3a was also observed in M12.4 cells (Figure2). The Diseases 2017, 5, 34 6 of 16 super-shift/ablation of the NF-µNR complex with anti-Cux conﬁrmed that the super-shifted complex withDiseases anti-Arid3a 2017, 5, 3410.3390/diseases5040034 is the NF-µNR complex. 6 of 16

Figure 3.3. Arid3aArid3a and and NF- NF-μµNRNR complexes complexes with with promoter-associated MARs are composed of 32 heterogeneous proteins.proteins. BCLBCL11 nuclear extract was prepared andand subjectedsubjected toto EMSAEMSA usingusing 32P-labeled Bf150 oror Tx125Tx125 fragments fragments as as probes. probes. Super-shift Super-shift assays assays were were performed performed with with antisera antisera or mAb or (mAbα) listed (α) abovelisted above the corresponding the corresponding lanes. Planelanes. arrowsPlane arrows indicate indicate the mobility the mobility of the NF-of theµNR NF- complex;μNR complex; arrows witharrows balls, with the balls, Arid3a the complex.Arid3a complex. No NE: No free NE: probe, free PI: probe, pre-immune PI: pre-immune serum. serum.

To furtherfurther testtest whether whether the the NF- NF-µNRμNR complex complex in in B cellsB cells contains contains Arid3a, Arid3a, we performedwe performed super-shift super- assaysshift assays using using nuclear nuclear extract extract prepared prepared from Cos-7 from cells,Cos-7 which cells, bywhich immunostaining by immunostaining and Western and Western blotting 1 (datablotting not (data shown) not containshown) nocontain detectable no detectable Arid3a protein Arid3a (Figure protein4 A).(Figure The Cos-74A). The and Cos-7 BCL 1andpatterns BCL canpatterns be compared can be compared in Figure in4B. Figure Even 4B. though Even increasingthough increasing amounts amounts of anti-Arid3a of anti-Arid3a clearly super-shiftedclearly super- shifted the NF-μNR complex in BCL1 B cells, it produced no effect on the NF-μNR complex in Cos-7 the NF-µNR complex in BCL1 B cells, it produced no effect on the NF-µNR complex in Cos-7 cells. cells. However, anti-Cux super-shifted the NF-μNR complex in both BCL1 and Cos-7. However, anti-Cux super-shifted the NF-µNR complex in both BCL1 and Cos-7. The presencepresence of Arid3aArid3a inin thethe NF-NF-µμNRNR complexcomplex inin BB cellscells waswas confirmedconfirmed byby two-dimensionaltwo-dimensional electrophoresis, employing employing EMSA EMSA in in the the first first dime dimensionnsion and and SDS-PAGE SDS-PAGE in in the second (Figure 44D).D). Western blotting blotting with with anti-Arid3a anti-Arid3a detected detected two two spots spots on the on membrane, the membrane, one corresponding one corresponding to its mobility to its mobilitywithin the within NF-μNR the complex NF-µNR and complex the other and to the its other mobility to itswithin mobility the Arid3a within complex the Arid3a (Figure complex 4D). (FigureTherefore,4D). we Therefore, concluded we that concluded Arid3a is thata component Arid3a is of a NF- componentμNR in B of cells NF- butµNR not inin non-B B cells cells. but not in non-BNext, cells. we addressed whether Arid3a is the only B cell-specific component required for this effect. We utilizedNext, we a nuclear addressed extract whether prepared Arid3a from isthe the Arid only3a retrovirally B cell-specific infected component stable cell required line (NIH3T3- for this effect.Arid3a; We described utilized in a nuclearMaterials extract and Methods). prepared fromAs shown the Arid3a in Figure retrovirally 4C, anti-Arid3a infected stableablated/super- cell line (NIH3T3-Arid3a;shifted both the Arid3a described and in NF- SectionμNR2 ).complexes. As shown inThis Figure indicated4C, anti-Arid3a that Arid3a ablated/super-shifted expression per se is bothsufficient the Arid3a for the and integration NF-µNR of complexes. Arid3a into This the indicatedNF-μNR complex. that Arid3a expression per se is sufficient for the integrationWe utilized of Arid3aseveral intoantibodies the NF- µtoNR check complex. the Arid3a and NF-μNR complexes for additional components.We utilized Both several BCL1 and antibodies M12.4 also to express check thethe B Arid3a cell-specific and NF- AridµNR protein, complexes Arid3b for[20]. additional Thus, we examined its contribution to the two EMSA complexes using an anti-Arid3b antiserum whose lack of components. Both BCL1 and M12.4 also express the B cell-specific Arid protein, Arid3b [20]. Thus, we examinedcross-reactivity its contribution against either to the Arid3a two EMSAor CDP/Cux complexes was usingpreviously an anti-Arid3b confirmed antiserum(data not shown). whose lack As ofshown cross-reactivity in Figures 3 against and 4, anti-Arid3b either Arid3a antiserum or CDP/Cux had no was effect previously on the NF- confirmedμNR complex (data notbut shown).slightly Asperturbed shown inthe Figures Arid3a3 andcomplex4, anti-Arid3b in BCL1. However, antiserum anti-Arid3b had no effect had on no the discernible NF- µNR complexeffect onbut the slightlyArid3a complex in M12.4 cells (Figures 3 and 4).

Diseases 2017, 5, 34 7 of 16

perturbed the Arid3a complex in BCL1. However, anti-Arid3b had no discernible effect on the Arid3a Diseases 2017, 5, 3410.3390/diseases5040034 7 of 16 complex in M12.4 cells (Figures3 and4).

µ Figure 4. TheThe NF- NF-μNRNR complex complex contains contains Arid3a Arid3a in inB cells B cells but but not not in non-B in non-B cells cells that that were were tested. tested. (A) µ (NF-A) NF-μNR NRin Cos-7 in Cos-7 includes includes CDP/Cux CDP/Cux but but not not Arid3a. Arid3a. Extract Extract and and probe probe preparation, preparation, EMSA and supershifts were performed as in Figure3. Triangles indicate increasing concentration of indicated supershifts were performed as in Figure 3. Triangles indicate increasing concentration of indicated antibodies. Arrows denote mobility of the NF-µNR complex; (B) Anti-Arid3a antiserum super-shifted antibodies. Arrows denote mobility of the NF-μNR complex; (B) Anti-Arid3a antiserum super-shifted the NF-µNR complex (plain arrows) as well as the Arid3a complex (arrows with balls) in BCL1; the NF-μNR complex (plain arrows) as well as the Arid3a complex (arrows with balls) in BCL1; (C) (C) The NF-µNR complex in NIH3T3 cells (NIH3T3-Arid3a) stably expressing Arid3a via retroviral The NF-μNR complex in NIH3T3 cells (NIH3T3-Arid3a) stably expressing Arid3a via retroviral infection; (D) Conﬁrmation of Arid3a within both NF-µNR and Arid3a EMSA complexes by 2-D gel infection; (D) Confirmation of Arid3a within both NF-μNR and Arid3a EMSA complexes by 2-D gel electrophoresis in B cells. First dimension, standard EMSA of nuclear extract prepared from M12.4. electrophoresis in B cells. First dimension, standard EMSA of nuclear extract prepared from M12.4. Second dimension, SDS-PAGE, was followed by Western blotting with anti-Arid3a. Positions of the Second dimension, SDS-PAGE, was followed by Western blotting with anti-Arid3a. Positions of the two EMSA complexes are indicated by arrows. Molecular markers are shown on the right of the gel. two EMSA complexes are indicated by arrows. Molecular markers are shown on the right of the gel.

A previous study [[27]27] indicated that Bruton tyrosinetyrosine kinase (Btk) interacts with Arid3a and is a component ofof thethe Arid3a-DNAArid3a-DNA complex. complex. However, However, we we were were unable unable to detectto detect super-shifts super-shifts ofeither of either the Arid3athe Arid3a or the or NF-the µNF-NRμ complexNR complex with with a commercially a commercially obtained obtain anti-Btked anti-Btk antibody antibody (Figure (Figure2). 2). A fraction of Arid3a Arid3a localizes localizes in in Hela Hela cells cells to to PML-nuclear PML-nuclear bodies bodies [28], [28 ],and and such such localization localization is isoften often accompanied accompanied by bySUMO-1 SUMO-1 post-t post-translationalranslational modificati modiﬁcation.on. Therefore, Therefore, we tested we tested whether whether an anti- an anti-SUMO-1SUMO-1 monoclonal monoclonal antibody antibody (provided (provided by byDr. Dr. G. G.Maul) Maul) could could super-shift super-shift the the Arid3a Arid3a complex. complex. Unexpectedly, anti-SUMO-1 super-shifted the NF-μµNR complex rather than the Arid3a complex (Figure3 3).). CDP/Cux and and Arid3b Arid3b were were reported reported to interact to interact with withRb [20,29]. Rb [20 Thus,,29]. we Thus, examined we examined the existence the existenceof Rb in the of NF- Rbμ inNR the or NF- Arid3aµNR complexes. or Arid3a Anti-Rb complexes. antibody Anti-Rb did antibodynot shift either did not of the shift complexes either of the(Figure complexes 3). Negative (Figure super-shift3). Negative controls, super-shift including controls, pre-immune including (PI) serum pre-immune and an (PI) irrelevant serum (anti- and anBCL11) irrelevant antiserum, (anti-BCL11) had no antiserum, effect on hadeither no co effectmplex. on These either super-shift complex. Theseassays super-shift demonstrate assays an demonstrateunexpected, anpreviously unexpected, unappreciated previously unappreciated complexity in complexity the NF-μ inNR the “repressive-” NF-µNR “repressive-” and Arid3a and Arid3a“activation-related” “activation-related” complexes complexes formed formedon the V onH1 the promoter-associated VH1 promoter-associated MARs. MARs.

3.2. DNA Binding Activity of Arid3a Isis Se Sensitivensitive to Cell Cycle and Nuclear Localization We investigatedinvestigated the the DNA DNA binding binding activity activity of theof the Arid3a Arid3a complex complex and theand NF- theµ NRNF- complexμNR complex under variousunder various conditions conditions such as such cell cycleas cell arrest cycle and arrest starvation. and starvation. Initially, Initially, we examined we examined the DNA the binding DNA binding affinity of the complexes after release of BCL1 cells from an aphidicolin-arrested cell cycle (see Materials and Methods for details). Nuclear extracts were prepared at two-hour intervals following release from G1/S arrest and were compared by EMSA to extracts prepared from asynchronous cells using Tx125 and Bf150 as probes. Following 11 hr after release, the Arid3a

Diseases 2017, 5, 34 8 of 16

afﬁnity of the complexes after release of BCL1 cells from an aphidicolin-arrested cell cycle (see Section2 Diseases 2017, 5, 3410.3390/diseases5040034 8 of 16 for details). Nuclear extracts were prepared at two-hour intervals following release from G1/S arrest and were compared by EMSA to extracts prepared from asynchronous cells using Tx125 and Bf150 as complex increased (relative to control) to a maximum at 4 h, while during the same interval, the NF- probes. Following 11 h after release, the Arid3a complex increased (relative to control) to a maximum μNR complex decreased (Figure 5A). The BCL1 cell cycle was monitored during the same time course at 4 h, while during the same interval, the NF-µNR complex decreased (Figure5A). The BCL 1 cell by propidium iodide staining and flow cytometry (Figure 5B). At 1–2 hr following release, most cells cycle was monitored during the same time course by propidium iodide staining and ﬂow cytometry were still in G1. At four hours, about approximately half of the cells remained in G1 phase and about (Figure5B). At 1–2 h following release, most cells were still in G 1. At four hours, about approximately half were in S phase. G2/M cells began to appear at 6 h, whereas the majority of the cells separated half of the cells remained in G1 phase and about half were in S phase. G2/M cells began to appear into G1 and G2/M at ~11 hr. The increase in the abundance of the Arid3a complex and the decrease in at 6 h, whereas the majority of the cells separated into G1 and G2/M at ~11 h. The increase in the theabundance NF-μNR complex of the Arid3a coincide complex with and the the S phase decrease maxima in the at NF- fourµNR hours complex following coincide release with thefrom S phase the cell cyclemaxima block. at four hours following release from the cell cycle block.

FigureFigure 5. 5.Arid3aArid3a binding binding to to promoter-associated promoter-associated MARs increasesincreases during during S S phase. phase. BCL BCL1 cell1 cell cycle cycle was was arrestedarrested with with 5 5μµg/mLg/mL aphidicolin aphidicolin forfor 1616 h. hours. Cells wereCells washed were washed to release to the release cell cycle the block,cell cycle cultured block, culturedfor 11 hfor with 11 aliquots hr with harvested aliquots every harvested 2 h (A )every Nuclei 2 were hr ( extractedA) Nuclei and were EMSA extracted was performed. and EMSA 0–11 was h performed.after cell cycle0–11: release, hours A:after asynchronous cell cycle release, cells B: A: super-shift asynchronous with anti-Arid3a cells B: super-shift antiserum, with C: super-shift anti-Arid3a antiserum,with anti-Cux C: super-shift antiserum, with N: freeanti-Cux probe. antiserum, Probes are N: indicated free probe. at the Prob top.es Plainare indicated arrows, NF-at theµNR, top. and Plain arrows with balls, Arid3a complexes; (B) BCL cells distribution at each time point based on the DNA arrows, NF-μNR, and arrows with balls, Arid3a1 complexes; (B) BCL1 cells distribution at each time pointcontent. based Cells on the were DNA ﬁxed content. with ethanol Cells were and fixed stained with with ethanol propidium and stained iodide. with Cell propidium cycle status iodide. was examined through FACS analysis; (C) Arid3a expression in BCL1 cells at each time point after the Cell cycle status was examined through FACS analysis; (C) Arid3a expression in BCL1 cells at each release of cell cycle block. Extracts were separated on SDS-PAGE and Western blotting was performed. time point after the release of cell cycle block. Extracts were separated on SDS-PAGE and Western (Loading controls not shown). blotting was performed. (Loading controls not shown).

We next determined the expression levels and the localization of Arid3a in these cells (Figure5C). We next determined the expression levels and the localization of Arid3a in these cells (Figure 5C). After release from cycle block, the expression of Arid3a appeared to increase in both the cytoplasm After release from cycle block, the expression of Arid3a appeared to increase in both the cytoplasm and the nucleus. However, we consistently observed that the Arid3a levels decreased at the six-hour and the nucleus. However, we consistently observed that the Arid3a levels decreased at the six-hour time point in both compartments, followed by a gradual, continued increase after 11 h. In addition, time point in both compartments, followed by a gradual, continued increase after 11 h. In addition, the nuclear/cytoplasmic ratio of Arid3a modestly increased at 4 h, post arrest, compared to the ratio thein nuclear/cytoplasmic the asynchronous culture. ratio of Therefore, Arid3a modestly DNA binding increased activity at 4 ofh, thepost Arid3a arrest, complex compared appeared to the ratio to in increasethe asynchronous as a result from culture. its increase Therefore, in nuclear DNA abundance.binding activity The data of indicatethe Arid3a that complex this increase appeared may be to increaseachieved as a by result increased from Arid3a its increase transcription, in nuclear increased abundance. nuclear The entry, data decreased indicate that nuclear this export increase or bymay be bothachieved mechanisms. by increased Arid3a transcription, increased nuclear entry, decreased nuclear export or by both mechanisms. 3.3. LPS Stimulates Formation of the NF-µNR Complex 3.3. LPS Stimulates Formation of the NF-μNR Complex Normal B cells and the BCL1 cell line respond to the B cell mitogen LPS by differentiating to a higherNormal antibody B cells secretion and the state BCL [9].1 cell Thus, line we respond examined to the the effect B cell of mitogen LPS on the LPS abundance by differentiating of the Arid3a to a higherand theantibody NF-µNR secretion EMSA complexes. state [9]. InThus, BCL 1wecells exam treatedined with the LPSeffect for of 2 days,LPS on we observedthe abundance an increase of the µ Arid3ain the and Arid3a the NF- complexμNR EMSA on both complexes. Bf150 and In Tx125 BCL MARs,1 cells treated whereas with the LPS NF- forNR 2 days, complex we observed decreased an increase in the Arid3a complex on both Bf150 and Tx125 MARs, whereas the NF-μNR complex decreased (Figure 6A). However, no changes were observed in anti-Arid3a, anti-Cux or anti-Arid3b super-shift patterns in either mock-treated or LPS-treated cells (Figure 6A). This prompted an examination of the expression and the localization of Arid3a and Cux within these cells. As shown in Figure 6B, LPS-treated cells expressed more Arid3a, but the nuclear/cytoplasmic ratio of Arid3a was unaffected. On the other hand, Cux accumulated only within the nucleus, and LPS treatment reduced

Diseases 2017, 5, 34 9 of 16

(Figure6A). However, no changes were observed in anti-Arid3a, anti-Cux or anti-Arid3b super-shift patterns in either mock-treated or LPS-treated cells (Figure6A). This prompted an examination of the expressionDiseases 2017, 5, and 3410.3390/diseases5040034 the localization of Arid3a and Cux within these cells. As shown in Figure9 of 16 6B, LPS-treated cells expressed more Arid3a, but the nuclear/cytoplasmic ratio of Arid3a was unaffected. On theits expression other hand, level Cux there. accumulated We conclude only that within thes thee mitogen-induced nucleus, and LPS reciproc treatmental changes reduced in Arid3a its expression and levelNF- there.μNR We DNA conclude can be attributed that these to mitogen-induced the concomitant changes reciprocal in the changes nuclear in abundance Arid3a and of NF-Arid3aµNR and DNA can beCux/CDP. attributed to the concomitant changes in the nuclear abundance of Arid3a and Cux/CDP.

FigureFigure 6. 6.Increased Increased Arid3a expression expression correlated correlated with with the increased the increased Arid3a-DNA Arid3a-DNA interaction. interaction. (A,B)

(A,BNuclei) Nuclei of ofLPS- LPS- or ormock-treated mock-treated BCL BCL1 cells1 cells were were extracted. extracted. (A) The (A) nuclear The nuclear extract extract was subjected was subjected to to EMSAEMSA with with Bf150 Bf150 probeprobe andand Tx125 prob probe.e. The The arrows arrows indicate indicate the the NF- NF-μNRµNR complex complex and and the thearrows arrows withwith a ball a ball indicate indicate the the Arid3a Arid3a complex; complex; ( B(B)) 22 µμgg or 5 μµgg of of extract extract were were separated separated on onSDS-PAGE SDS-PAGE and and Western blotting was performed using anti-Arid3a antiserum or anti-Cux antiserum; (C,D) BCL1 cells Western blotting was performed using anti-Arid3a antiserum or anti-Cux antiserum; (C,D) BCL1 cells werewere grown grown with with or without or without fresh fresh media media supplement supplement for 3for days. 3 days. Nuclei Nuclei were were extracted extracted (C) The(C) The nuclear nuclear extract was subjected to EMSA with Bf150 probe; (D) 2 μg of extract was separated on SDS- extract was subjected to EMSA with Bf150 probe; (D) 2 µg of extract was separated on SDS-PAGE and PAGE and Western blotting was performed with anti-Arid3a antiserum. Western blotting was performed with anti-Arid3a antiserum. 3.4. Increased Nuclear Levels of Arid3a is Sufficient for High Affinity Arid3a-MAR Complex Formation 3.4. Increased Nuclear Levels of Arid3a Is Sufficient for High Affinity Arid3a-MAR Complex Formation Starvation of cultured mammalian cells results in a mid-G1 arrest (quiescence) that differs mechanisticallyStarvation of culturedfrom G1/S mammalian or G0 arrestcells (senescence) results in[30]. a mid-GTo de1terminearrest the (quiescence) potential effect that differsof mechanisticallystarvation on from Arid3a G1 /SDNA or Gbinding,0 arrest BCL (senescence)1 cells were [30 allowed]. To determine to reach theconfluence potential by effect growing of starvation them on Arid3afor 3 days DNA without binding, media BCL supplemen1 cells weretation. allowed Nuclear to extracts reach were confluence prepared by and growing subjected them to EMSA. for 3 days withoutAs shown media in supplementation. Figure 6C, starvation Nuclear abolished extracts the Arid3a-DNA were prepared complex and subjectedand slightly to increased EMSA. As the shown NF- in FigureμNR6C, complex. starvation In abolishedstarved cells, the Arid3a Arid3a-DNA expression complex was reduced and slightly as compared increased with the that NF- inµ regularlyNR complex. In starvedsupplemented cells, Arid3a cells. expressionHowever, there was reducedwas no signific as comparedant change with in that the innuclear/cytoplasmic regularly supplemented ratio of cells. Arid3a (Figure 6D). However, there was no significant change in the nuclear/cytoplasmic ratio of Arid3a (Figure6D). Cell cycle arrest, LPS stimulation and starvation experiments together indicated that increased Cell cycle arrest, LPS stimulation and starvation experiments together indicated that increased nuclear Arid3a expression correlates with increased abundance of the Arid3a-promoter MAR EMSA nuclearcomplex Arid3a and expression reduced abundance correlates withof the increased NF-μNR abundancecomplex assembled of the Arid3a-promoter on the promoter MAR MARs. EMSA complexTherefore, and reduced we asked abundance whether merely of the NF- increasingµNR complex the levels assembled of Arid3a on is the sufficient promoter by MARs.performing Therefore, a we askedsimple whether mixing merelyexperiment increasing (Figure the7). levelsFixed ofamounts Arid3a of is BCL sufficient1 nuclear by performingextract werea mixed simple with mixing experimentincreasing (Figure amounts7). Fixedof in vitro amounts translated of BCL Arid3a1 nuclear, and EMSAs extract were were performe mixedd with using increasing Tx125 as probe. amounts of inIncreasing vitro translated in vitro Arid3a,translated and Arid3a EMSAs abrogated were performed the NF-μNR using complex Tx125 while as probe. increasing Increasing the Arid3ain vitro translatedcomplex Arid3a (Figure abrogated 7). Therefore, the we NF- suggestµNR complexthat an increased while increasing amount of Arid3a the Arid3a in the complex nucleus may (Figure be 7). Therefore,all that weis required suggest for that high an increasedaffinity Arid3a-MAR amount of complex Arid3a information. the nucleus may be all that is required for high affinity Arid3a-MAR complex formation.

Diseases 2017, 5, 34 10 of 16 Diseases 2017, 5, 3410.3390/diseases5040034 10 of 16

Figure 7. Arid3a over-expression abrogates NF-μNR-DNA binding. BCL1 nuclei were extracted and Figure 7. Arid3a over-expression abrogates NF-µNR-DNA binding. BCL1 nuclei were extracted and the extractthe extract was was mixed mixed with within in vitro vitrotranslated translated Arid3aArid3a before before DNA DNA binding. binding. The The mixture mixture was was subject subject to EMSA using Tx125 probe. The unmodified arrows indicate the NF-μNR complex and the dotted to EMSA using Tx125 probe. The unmodified arrows indicate the NF-µNR complex and the dotted arrows indicate the Arid3a complex. arrows indicate the Arid3a complex. 3.5. Transactivation Activity of Arid3a in Non-B Cells is Modulated by Promoter-Associated MARs 3.5. Transactivation Activity of Arid3a in Non-B Cells Is Modulated by Promoter-Associated MARs A previous report demonstrated that a rearranged S107 IgH transgene driven by the VH1 core promoterA previous (−125 report bp upstream) demonstrated devoid thatof MAR a rearranged regions is S107not expressed IgH transgene in transgenic driven non-lymphoid by the VH1 core promotercells [12]. (−125 The bp authors upstream) concluded devoid from ofMAR these regionsdata that is notsequences expressed −125 inbp transgenic upstream non-lymphoidof the VH1 cellstranscription [12]. The authorsstart site concludedare sufficient from for the these repressi dataon that of an sequences IgH gene −in125 non-B bp cells. upstream However, of thethat VH1 transcriptionresult in no start way siteformally are sufficient excluded forthe thepossibility repression that the of anpromoter-associat IgH gene in non-Bed MARs cells. upstream However, to - that result125 in repress no way non-B formally IgH transcription excluded the in the possibility absence thatof an the enhancer; promoter-associated i.e., the situation MARs achieved upstream in to −“germline”125 repress targeting non-B IgHof rearrangement transcription into thea particular absence V ofH ansegment enhancer; [9]. To i.e., test the the situation effect of achieved the promoter-associated MARs (5’ MARs), we generated firefly luciferase (pGL3-promoter) vectors in “germline” targeting of rearrangement to a particular VH segment [9]. To test the effect of the containing Bf150 (pGL3bfp) and Bf150 plus Tx125 (pGL3btp) upstream of the SV40 minimal core promoter-associated MARs (5’ MARs), we generated firefly luciferase (pGL3-promoter) vectors promoter (Figure 8). To test the effect of the enhancer, we constructed Eμ containing reporter containingplasmids Bf150 (pGL3pE (pGL3bfp)μ and pGL3btpE and Bf150μ) in plus which Tx125 Eμ was (pGL3btp) inserted downstream upstream of of the the SV40poly-A minimal signal of core promoterthe reporter (Figure gene.8). To These test were the effect then oftransiently the enhancer, transfected we constructed into NIH3T3 E withµ containing each of the reporter pGL3 firefly plasmids (pGL3pEluciferaseµ and reporter pGL3btpE plasmidsµ) in whichalong with Eµ was pRL-Renilla inserted downstreamas loading controls. of the poly-AAfter 48 signal h, the of luciferase the reporter gene.activity These was were measured then transiently and the transactivation transfected intoactivity NIH3T3 was analyzed. with each As ofshown the pGL3in Figure firefly 8A, luciferaseBf150 reporterplus Tx125 plasmids (5’ MARs) along as with well pRL-Renilla as Eμ and 5’ as MARs loading or E controls.μ alone repressed After 48 reporter h, the luciferase gene transcription activity was measuredin NIH3T3 and cells thetransactivation to levels varying activity from ~3–20 was analyzed.fold, while As Bf150 shown alone in Figurehad no8 A,effect. Bf150 These plus data Tx125 (5’ MARs)indicated as that well Bf150 as E µplusand Tx125 5’ MARs may mediate or Eµ alone the repression repressed of reporterIgH gene genetranscription transcription in non-B in cells. NIH3T3 cellsHowever, to levels varyinga significant from de-repression ~3–20 fold, while(~7 fold) Bf150 of the alone Eμ-mediated had no effect. effect These was detected data indicated when E thatμ and Bf150 plus5’ Tx125 MARs may were mediate positioned the upon repression the same of IgH construct. gene transcription in non-B cells. However, a significant To examine the effect of Arid3a on these regulatory elements, we performed the same de-repression (~7 fold) of the Eµ-mediated effect was detected when Eµ and 5’ MARs were positioned transfections in NIH3T3-Arid3a cells (NIH3T3 cells stably expressing retroviral Arid3a). While a upon the same construct. similar pattern of repression was observed, the magnitudes of several of the reporters were To examine the effect of Arid3a on these regulatory elements, we performed the same transfections significantly reduced, indicative of specific de-repression by Arid3a (data not shown). To best in NIH3T3-Arid3aillustrate this, we cells compared (NIH3T3 the cells luciferase stably activities expressing obtained retroviral in NIH3T3 Arid3a). to Whilethose from a similar NIH3T3- pattern of repressionArid3a cells was (Figure observed, 8B). Arid3a the magnitudes expression increased of several the of basal the level reporters transcription were significantly through the SV40 reduced, indicativeminimal of promoter. specific de-repression As expected from by Arid3a the previous (data notresults, shown). Arid3a To had best no illustrate effect on this, the weluciferase compared the luciferaseactivity mediated activities through obtained Bf150 in when NIH3T3 basal to thoselevel increase from NIH3T3-Arid3a was compensated. cells However, (Figure8 B).Arid3a Arid3a expressionexpression increased moderately the increased basal level the transcription reporter gene throughtranscription the SV40mediat minimaled through promoter. 5’ MARs, As Eμ expected-only fromor the both previous elements results, combined. Arid3a As had we no previously effect on the observed luciferase in activityB cells mediated[10], the highest through levels Bf150 of when basal level increase was compensated. However, Arid3a expression moderately increased the reporter gene transcription mediated through 5’ MARs, Eµ-only or both elements combined. As we previously Diseases 2017, 5, 34 11 of 16 Diseases 2017, 5, 3410.3390/diseases5040034 11 of 16 transactivationobserved in B activity cells [10 ],were the highestachieved levels when of transactivationthe 5′ MARs and activity Eμ together were achieved flank the when core the promoter 50 MARs on theand same Eµ vectortogether (Figure flank the8B). core promoter on the same vector (Figure8B). Collectively,Collectively, these these experiments experiments indicate indicate that Arid3aArid3a cancan transactivatetransactivate IgH IgH gene gene transcription transcription throughthrough linked linked 5′ 5MARs0 MARs and and E Eμµ inin non-B non-B cells cells by de-repressingde-repressing the the locus. locus.

FigureFigure 8. 8.Arid3aArid3a transactivates transactivates the corecoreSV40 SV40 promoter promoter in in non-B non-B cells cells through through promoter-associated promoter-associated (5’) (5’)MARs MARs (Bf150 (Bf150 and and Tx125). Tx125). NIH3T3 NIH3T3 cells andcells NIH3T3-Arid3a and NIH3T3-Arid3a cells (which cells express(which retroviral express Arid3a)retroviral Arid3a)were transfectedwere transfected with firefly with luciferase firefly luciferase vectors containing vectors containing Bf150, 5’ MARs Bf150, (Bf150 5’ MARs + Tx125), (Bf150 Eµ or+ Tx125), 5’ MAR Eμ or +5’ EMARµ and + RenillaEμ and luciferase Renilla luciferase as control. as At control. 24 h after At 24 transfection, h after transfection, cell lysates cell were lysates prepared were and prepared the andluciferase the luciferase activities activities were measured were measured and normalized and normalized using the using dual the luciferase dual luciferase system. ( Asystem.) 5’ MARs (A) 5’ MARsrepress repress SV40 promoter-drivenSV40 promoter-driven transcription transcription in NIH3T3 in cells. NIH3T3 Values cells. are plotted Values relative are plotted to 100% relative for the to µ 100%control for the (construct control p); (construct (B) Arid3a p); transactivates (B) Arid3a transactivates the SV40 core the promoter SV40 core through promoter 5’ MARs through and/or 5’ E MARsin NIH3T3-Arid3a. Values are obtained from a NIH3T3-Arid3a cell line as above and divided by the values and/or Eμ in NIH3T3-Arid3a. Values are obtained from a NIH3T3-Arid3a cell line as above and (shown in A) obtained from NIH3T3 and plotted relative to 100% for the control. Values were obtained divided by the values (shown in A) obtained from NIH3T3 and plotted relative to 100% for the control. from three independent experiments. (C) Graphical summary of reporter gene constructs used. Values were obtained from three independent experiments. (C) Graphical summary of reporter gene constructs used. 4. Discussion 4. DiscussionTissue-specific transcription of a progressively rearranged IgH gene has been shown to require a core VH promoter and the intronic enhancer Eµ [7]. Eµ has been shown to be sufficient for the repression Tissue-specific transcription of a progressively rearranged IgH gene has been shown to require of IgH transcription in non-B cells. In this study, we showed that the VH1 promoter-associated MARs, a coreBf150 V andH promoter Tx125, also and can the repress intronic reporter enhancer gene E transcriptionμ [7]. Eμ has in been non-B shown cells in to a be concerted sufficient fashion for the repressionand, further, of IgH they transcription can relieve the in repressionnon-B cells. mediated In this bystudy, Eµ in we non-B showed cells. that Arid3a the transactivated VH1 promoter- associatedreporter MARs, gene transcription Bf150 and throughTx125, also Eµ or can the repress 5’ MARs. reporter Arid3a gene transactivation transcription was in maximal non-B cells when in a concertedEµ and thefashion 5’ MARs and, flankingfurther, thethey promoter can relieve were the positioned repression on mediated the same construct.by Eμ in non-B This is cells. a context Arid3a transactivatedanalogous to reporter that following gene transcription VDJ rearrangement, through in E whichμ or the the 5’ promoter MARs. Arid3a and Eµ transactivationare juxtaposed towas maximalfunction when synergistically Eμ and the for 5’ IgH MARs gene flanking transcription the promoter in B cells. were positioned on the same construct. This is aBf150 context and analogous Tx125 are occupiedto that following by the NF- VDJµNR rearrangement, and Arid3a complexes in which in the B cells, promoter but only and by E theμ are juxtaposedNF-µNR complexto function in non-B synergistically cells. Here, for we IgH showed gene that transcription Arid3a is a componentin B cells. of the NF-µNR complex asBf150 well as and the Tx125 Arid3a are complex occupied in Bby cells. the NF- Arid3a,μNR when and Arid3a assembled complexes into the in NF- Bµ cells,NR complex, but only mayby the NF-conferμNR differentcomplex functionin non-B to cells. this complex Here, we in Bshowed cells and that non-B Arid3a cells. is One a component hypothesis isof thatthe Arid3aNF-μNR complexmay relieve as well the as repression the Arid3a mediated complex by in NF-B cells.µNR—an Arid3a, idea when supported assembled by the into transactivation the NF-μNR datacomplex, of mayFigure confer8. Anotherdifferent possibility function to is thatthis Arid3acomplex may in B weaken cells and the non-B NF- µNR-DNA cells. One interaction hypothesis and is that facilitate Arid3a mayrapid relieve dissociation the repression of the complex mediated in B by cells, NF- particularlyμNR - an idea in response supported to stimulation by the transactivation with agents such data as of LPS. Because Arid3a expression, in itself, was sufficient for its incorporation into the NF-µNR-DNA Figure 8. Another possibility is that Arid3a may weaken the NF-μNR-DNA interaction and facilitate complex in non-B cells, Arid3a participation may be prerequisite for the de-repression of IgH prior to rapid dissociation of the complex in B cells, particularly in response to stimulation with agents such the time in which it might actively accelerate transcriptional initiation in a positive manner. as LPS. Because Arid3a expression, in itself, was sufficient for its incorporation into the NF-μNR- DNA complex in non-B cells, Arid3a participation may be prerequisite for the de-repression of IgH prior to the time in which it might actively accelerate transcriptional initiation in a positive manner. Our EMSA super-shift data indicated that the NF-μNR complex contains SUMO-1 modified protein(s) both in B and non-B cells. In many cases, SUMO-1 modification is required for sub-nuclear

Diseases 2017, 5, 34 12 of 16

Our EMSA super-shift data indicated that the NF-µNR complex contains SUMO-1 modified protein(s) both in B and non-B cells. In many cases, SUMO-1 modification is required for sub-nuclear localization such as within the PML-nuclear body [31]. This raises the possibility that the promoter-associated MARs, when bound by NF-µNR, may become sequestered into a restricted, heterochromatin region, such as the PML body, in non-B cells. This would lead to a repressed state of IgH. Indeed, several studies have shown that genes can change their nuclear neighborhoods by re-positioning from repressive sites to transcriptionally active compartments, and vice versa [32–35]. The nuclear positionings of Ig loci also are tightly regulated, as they are positioned centrally in committed B-cell progenitors, while in non-B cells they are located at the nuclear periphery [35–37]. It is conceivable that Arid3a, via its ability to condense chromatin and partition nucleosomes, might contribute to such relocalization. Re-localization of the IgH locus from a sequestered nuclear area would facilitate the access of other transcription factors to the locus. The Arid3a EMSA complex also includes Arid3b. Previously, we showed that Arid3a binds Arid3b, a paralog of Arid3a, through a REKLES domain in vitro [22]. Therefore, we reasoned that Arid3b may exist in a form of homomers or Arid3a-Arid3b heteromers in the Arid3a/DNA complex. From the data of Figures2 and3, it would appear that most of the Arid3a complex in BCL 1 is composed of Arid3a-Arid3b heteromers, because both anti-Arid3a and anti-Arid3b quantitatively super-shift the complex. However, while Arid3b and CDP/Cux may be recruited to the Arid3a complex in BCL1, this does not appear to be the case in M12.4 B cells. Nor does either B cell line appear to contain Arid3b in the NF-µNR complex. These results suggest that the differential composition of the Arid3a complex may dictate differential functions or activities for Arid3a during B cell development and differentiation. However, we were unable to confirm previous results that predict that both Rb and Btk participate in the Arid3a and NF-µNR complexes, respectively [20,27]. It is unclear at this time whether this is a technical problem, such as gel or antibody conditions, a problem of cell heterogeneity or whether neither actually participates in the complexes. Additional studies are needed to resolve this issue. Wang et al. demonstrated that over-expression of CDP, a component of NF-µNR, abrogated Arid3a-MAR DNA interaction [15]. Conversely, we found that Arid3a over-expression can abrogate NF-µNR MAR DNA binding. This suggests a straightforward model in which the relative amount of Arid3a and NF-µNR available within the correct nuclear location determines the relative DNA binding activity of Arid3a and NF-µNR complexes. Transcription of the genes for the human histone proteins are activated at G1/S phase transition and maintained during S phase by HiNF-D, whose functional core component is Cux. Cux interacts with cell cycle control elements not only within HiNF-D but singularly in other systems [29]. We induced cell cycle arrest at G1/S using the DNA-polymerase α inhibitor, aphidicolin. Following release, BCL1 cells showed decreased DNA binding of NF-µNR and increased DNA binding of Arid3a. HiNF-D and NF-µNR consist of different components even though CDP/Cux is found in both. These differences may underlie the disparity of the DNA binding properties of the two complexes. The enhanced Arid3a DNA binding we observed at four hours following release appears to result directly from increased Arid3a expression. In addition, starvation decreased Arid3a expression and Arid3a-DNA binding but increased NF-µNR-DNA binding. Increased Arid3a expression following LPS stimulation in B cells also correlated with increased DNA binding of Arid3a and reduced DNA binding of NF-µNR. Therefore, we conclude that the Arid3a-NF-µNR ratio in the nucleus plays a crucial role in regulating IgH gene transcription in B cells and its repression in non-B cells. IgH gene transcription is regulated in a highly complicated manner. This process is controlled by a variety of transcription factors, and many of them, including Arid3a, are subject to regulation by their tissue specificity, development/differentiation stage specificity and extracellular stimulation. For example, LPS or IL-5 and antigen induced the differentiation of BCL1 cells to plasmablasts/plasma cells specialized for the secretion of high levels of antibodies [38,39] and increased Arid3a-MAR 0 binding (Figure6A) [ 40]. The VH1 promoter-associated 5 MARs, Bf150 and Tx125, were elucidated here as elements involved in the regulation of IgH gene transcription. Diseases 2017, 5, 34 13 of 16

We have summarized our data in the model of Figure9. When occupied by NF- µNR, the MARs function to repress IgH in non-B cells. From the presence of SUMO-1 in this complex, we hypothesized that the IgH VH-associated promoter region may be sequestered within a subnuclear region(s) in which most transcription activities are repressed, such as within nuclear dots (Figure9A). In B cells, the 5’ MARs can be bound by the NF-µNR and/or the Arid3a complex. The equilibrium between these complexesDiseases 2017 depends, 5, 3410.3390/diseases5040034 on the relative expression levels of Arid3a and CDP/Cux. These13 of 16 levels are controlledregion(s) in a dominant in which waymost transcription as Arid3a expression activities are levelsrepressed, increase. such as within But as nuclear Arid3a dots expression (Figure 9A). levels rise (e.g., in theIn transition B cells, the from5’ MARs a pre-B can be cell bound to aby mature the NF- BμNR cell and/or or following the Arid3a mitogenic complex. The stimulation equilibrium of mature B cells), Arid3abetween partitions these complexes into both depends the on NF- theµ NRrelative complex expression and levels the Arid3aof Arid3a complex. and CDP/Cux. The These interaction of Arid3a withlevels NF- areµ NRcontrolled may in change a dominant the DNAway as Arid3a binding expression affinity levels of the increase. NF-µ ButNR as complex Arid3a expression and/or increase levels rise (e.g., in the transition from a pre-B cell to a mature B cell or following mitogenic stimulation µ the accessibilityof mature of B other cells), componentsArid3a partitions which into both may the comprise NF-μNR complex the complex. and the TheseArid3a complex. changes The in NF- NR may alter theinteraction complex of Arid3a suchthat with itNF- canμNR dissociate may change from the theDNA MARs binding (Figure affinity9 B).of the In addition,NF-μNR complex the two MARs appear to haveand/ordifferent increase the affinities accessibility for theof other NF- µcomponentsNR complex which and may the comprise Arid3a complex.the complex. In These unstimulated cells, Tx125changes has high in NF- affinityμNR may for alter NF- theµ NR,complex while such Bf150 that it hascan dissociate high affinity from the for MARs the Arid3a (Figure 9B). complex. In This addition, the two MARs appear to have different affinities for the NF-μNR complex and the Arid3a could allowcomplex. the IgH In promoterunstimulated to cells, associate Tx125 has with high putative affinity for repressive NF-μNR, while sub-nuclear Bf150 has regionshigh affinity more for loosely in B cells thanthe in Arid3a non-B complex. cells. Finally, This could we allow suggest the IgH that promoter activation to associate of B cells with increases putative repressive the affinity sub- of Arid3a for both MARs,nuclear allowing regions more the loosely IgH promoter in B cells than to be in liberatednon-B cells. from Finally, the we repressive suggest that region activation by displacementof B of NF-µNRcells (Figure increases9C). the As affinity cells becomeof Arid3a morefor both activated MARs, allowing and move the IgH toward promoter terminal to be liberated differentiation, from the the repressive region by displacement of NF-μNR (Figure 9C). As cells become more activated and Arid3a complexmove toward may terminal change differentiation, its components the Arid3a so as complex to re-localize may change the IgHits components promoter so to as ato euchromatic re- sub-nuclearlocalize region. the IgH promoter to a euchromatic sub-nuclear region.

Figure 9. HypotheticalFigure 9. Hypothetical function function of Arid3a of Arid3a in in regulating regulating IgH IgH transcription transcription through through 5′ MARs. 50 (MARs.A) IgH (A) IgH VH-associated promoter region may be sequestered within a sub-nuclear region(s) in which most VH-associatedtranscription promoter activities region are repressed may be in sequestered non-B cells; (B) within The interaction a sub-nuclear of Arid3a with region(s) CDP/Cux in may which most transcriptionchange activities the DNA are binding repressed affinity in of non-B the NF- cells;μNR complex (B) The and/or interaction increase of the Arid3a accessibility with of CDP/Cux other may change thecomponents DNA binding which afﬁnitymay comprise of the the NF- Arid3aµNR complex. complex These and/or changes increase in NF-μ theNR accessibilitymay alter the of other componentscomplex which such may that comprise it can dissociate the Arid3a from MARs; complex. (C) Activation These changes of the B incells NF- increasesµNR may the affinity alter the of complex Arid3a for both MARs, allowing the IgH promoter to be liberated from the repressive region by such that it can dissociate from MARs; (C) Activation of the B cells increases the afﬁnity of Arid3a for displacement of NF-μNR; (D) As cells become more activated and move toward terminal both MARs,differentiation, allowing the the IgHArid3a promoter complex may to bechange liberated its components from the so repressiveas to re-localize region the IgH by promoter displacement of NF-µNR; (D) As cells become more activated and move toward terminal differentiation, the Arid3a complex may change its components so as to re-localize the IgH promoter to a euchromatic sub-nuclear

region; (E) Hypothesized equilibrium of Arid3a and NF-µNR during B cell development. Diseases 2017, 5, 34 14 of 16

Supplementary Materials: The following are available online. Figure S1: Intronic (Eµ) and 30 enhancers within the IgH locus. (A) Eµ is composed of a core enhancer and two flanking MAR regions. MAR regions contain four A/T-rich P-site protein binding motifs (hatched boxes) and the core enhancer contains several motifs (open boxes). Transcription factors are shown under the motifs to which they bind; (B) IgH 30 enhancer is comprised of four DNase I hypersensitive sites (open boxes and arrows). HS3a and HS4 are inverted repeats that form palindrome structures; (C) Position of MARs within the rearranged IgH locus. Closed ovals indicate MARs; open ovals, enhancers; open boxes, exons. Acknowledgments: We thank June Harriss for her excellent contribution to all aspects of the animal husbandry, Chhaya Das and Maya Ghosh for help in cell culture, and molecular techniques. We thank members of the Tucker laboratory for discussions and reading of the manuscript. Support for this work was provided by NIH Grant R01CA31534, Cancer Prevention Research Institute of Texas (CPRIT) Grants RP100612, RP120348; and the Marie Betzner Morrow Centennial Endowment to H.O.T. Author Contributions: D.K. and H.O.T. designed the research; D.K., C.S., and M.A.B. performed the research; D.K., C.S., M.A.B. and H.O.T. analyzed data; D.K. and H.O.T. wrote the manuscript; and D.K. and H.O.T. initiated the project. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Wasylyk, C.; Wasylyk, B. The immunoglobulin heavy-chain B-lymphocyte enhancer efficiently stimulates transcription in non-lymphoid cells. EMBO J. 1986, 5, 553–560. [PubMed] 2. Matthias, P.; Baltimore, D. The immunoglobulin heavy chain locus contains another B-cell-specific 3’ enhancer close to the alpha constant region. Mol. Cell. Biol. 1993, 13, 1547–1553. [CrossRef][PubMed] 3. Chauveau, C.; Jansson, E.A.; Muller, S.; Cogne, M.; Pettersson, S. Cutting edge: Ig heavy chain 3’ HS1–4 directs correct spatial position-independent expression of a linked transgene to B lineage cells. J. Immunol. 1997, 163, 4637–4641. 4. Le Noir, S.; Boyer, F.; Lecardeur, S.; Brousse, M.; Oruc, Z.; Cook-Moreau, J.; Denizot, Y.; Cogné, M. Functional anatomy of the immunoglobulin heavy chain 3’ super-enhancer needs not only core enhancer elements but also their unique DNA context. Nucleic Acids Res. 2017, 45, 5829–5837. [CrossRef][PubMed] 5. Cockerill, P.N.; Yuen, M.H.; Garrard, W.T. The enhancer of the immunoglobulin heavy chain locus is flanked by presumptive chromosomal loop anchorage elements. J. Biol. Chem. 1987, 262, 5394–5397. [PubMed] 6. Banerji, J.; Olson, L.; Schaffner, W. A lymphocyte-specific cellular enhancer is located downstream of the joinings region in immunoglobulin heavy chain genes. Cell 1983, 33, 729–740. [CrossRef] 7. Staudt, L.M. Immunoglobulin gene transcription. Annu. Rev. Immunol. 1991, 9, 373–398. [CrossRef] [PubMed] 8. Perez-Mutul, J.; Macchi, M.; Wasylyk, B. Mutational analysis of the contribution of sequence motifs within the IgH enhancer to tissue specific transcriptional activation. Nucleic Acids Res. 1998, 16, 6085–6096. [CrossRef] 9. Webb, C.F.; Zong, R.T.; Lin, D.; Wang, Z.; Kaplan, M.; Paulin, Y.; Smith, E.; Probst, L.; Bryant, J.; Goldstein, A.; et al. Differential regulation of immunoglobulin gene transcription via nuclear matrix-associated regions. Cold Spring Harb. Symp. Quant. Biol. 1999, 64, 109–118. [CrossRef][PubMed] 10. Kaplan, M.H.; Zong, R.T.; Herrscher, R.F.; Scheuermann, R.H.; Tucker, P.W. Transcriptional activation by a matrix associating region-binding protein. Contextual requirements for the function of bright. J. Biol. Chem. 2001, 276, 21325–21330. [CrossRef][PubMed] 11. Herrscher, R.F.; Kaplan, M.H.; Lelsz, D.L.; Das, C.; Scheuermann, R.; Tucker, P.W. The immunoglobulin heavy-chain matrix-associating regions are bound by Bright: A B cell-specific trans-activator that describes a new DNA-binding protein family. Genes Dev. 1995, 9, 3067–3082. [CrossRef][PubMed] 12. Avitahl, N.; Calame, K.A. 125 bp region of the Ig VH1 promoter is sufficient to confer lymphocyte-specific expression in transgenic mice. Int. Immunol. 1996, 8, 1359–1366. [CrossRef][PubMed] 13. Grosschedl, R.; Weaver, D.; Baltimore, D.; Costantini, F. Introduction of a mu immunoglobulin gene into the mouse germ line: Specific expression in lymphoid cells and synthesis of functional antibody. Cell 1988, 38, 647–658. [CrossRef] 14. Aguilera, R.J.; Hope, T.J.; Sakano, H. Characterization of immunoglobulin enhancer deletions in murine plasmacytomas. EMBO J. 1985, 4, 3689–3693. [PubMed] Diseases 2017, 5, 34 15 of 16

15. Wang, Z.; Goldstein, A.; Zong, R.T.; Lin, D.; Neufeld, E.J.; Scheuermann, R.H.; Tucker, P.W. Cux/CDP homeoprotein is a component of NF-µNR and represses the immunoglobulin heavy chain intronic enhancer by antagonizing the Bright transcription activator. Mol. Cell. Biol. 1999, 19, 284–295. [CrossRef][PubMed] 16. Neufeld, E.J.; Skalnik, D.G.; Lievens, P.M.; Orkin, S.H. Human CCAAT displacement protein is homologous to the Drosophila homeoprotein, cut. Nat. Genet. 1992, 1, 50–55. [CrossRef][PubMed] 17. Li, S.; Moy, L.; Pittman, N.; Shue, G.; Aufiero, B.; Neufeld, E.J.; LeLeiko, N.S.; Walsh, M.J. Transcriptiona l repression of the cystic fibrosis transmembrane conductance regulator gene, mediated by CCAAT displacement protein/cut homolog, is associated with histone deacetylation. J. Biol. Chem. 1999, 274, 7803–7815. [CrossRef][PubMed] 18. Coqueret, O.; Berube, G.; Nepveu, A. The mammalian Cut homeodomain protein functions as a cell-cycle-dependent transcriptional repressor which downmodulates p21WAF1/CIP1/SDI1 in S phase. EMBO J. 1998, 17, 4680–4694. [CrossRef][PubMed] 19. Van Gurp, M.F.; Pratap, J.; Luong, M.; Javed, A.; Hoffmann, H.; Giordano, A.; Stein, J.L.; Neufeld, E.J.; Lian, J.B.; Stein, G.S.; et al. The CCAAT displacement protein/cut homeodomain protein represses osteocalcin gene transcription and forms complexes with the retinoblastoma protein-related protein p107 and cyclin A. Cancer Res. 1999, 59, 5980–5988. [PubMed] 20. Numata, S.; Claudio, P.P.; Dean, C.; Giordano, A.; Croce, C.M. Bdp, a new member of a family of DNA-binding proteins, associates with the retinoblastoma gene product. Cancer Res. 1999, 59, 3741–3747. [PubMed] 21. Kim, D.; Tucker, P.W. A regulated nucleocytoplasmic shuttle contributes to Bright’s function as a transcriptional activator of immunoglobulin genes. Mol. Cell. Biol. 2006, 26, 2187–2201. [CrossRef] [PubMed] 22. Kim, D.; Probst, L.; Das, C.; Tucker, P.W. REKLES is an ARID3-restricted multifunctional domain. J. Biol. Chem. 2007, 25, 15768–15777. [CrossRef][PubMed] 23. Laskov, R.; Kim, J.K.; Woods, V.L.; McKeever, P.E.; Asofsky, R. Membrane immunoglobulins of spontaneous B-lymphomas of aged BALB/c mice. Eur. J. Immunol. 1981, 11, 462–471. [CrossRef][PubMed] 24. Gronowicz, E.S.; Doss, C.A.; Howard, F.D.; Morrison, D.C.; Strober, S. An in vitro line of the B cell tumor BCL1 can be activated by LPS to secrete IgM1. J. Immunol. 1980, 125, 976–980. [PubMed] 25. Johnson, D.R.; Levanat, S.; Bale, A.E. Isolation of intact nuclei for nuclear extract preparation from a fragile B-lymphocyte cell line. Bio-Techniques 1995, 19, 192–195. 26. Webb, C.F.; Das, C.; Eneff, K.L.; Tucker, P.W. Identification of a matrix associated region 5’ of an immunoglobulin heavy chain variable region gene. Mol. Cell. Biol. 1991, 11, 5206–5211. [CrossRef] [PubMed] 27. Webb, C.F.; Yamashita, Y.; Ayers, N.; Evetts, S.; Paulin, Y.; Conley, M.E.; Smith, E.A. The transcription factor Bright associates with Bruton’s tyrosine kinase, the defective protein in immunodeficiency disease. J. Immunol. 2000, 165, 6956–6965. [CrossRef][PubMed] 28. Zong, R.T.; Das, C.; Tucker, P.W. Regulation of matrix attachment region dependent, lymphocyte-restricted transcription through differential localization within promyelocytic leukemia nuclear bodies. EMBO J. 2000, 19, 4123–4133. [CrossRef][PubMed] 29. Van Wijnen, A.J.; van Gurp, M.F.; de Ridder, M.C.; Tufarelli, C.; Last, T.J.; Birnbaum, M.; Vaughan, P.S.; Giordano, A.; Krek, W.; Neufeld, E.J.; et al. CDP/cut is the DNA-binding subunit of histone gene transcription factor HiNF-D: A mechanism for gene regulation at the G1/S phase cell cycle transition point independent of transcription factor E2F. Proc. Natl. Acad. Sci. USA 1996, 93, 11516–11521. [CrossRef][PubMed] 30. Ho, A.; Dowdy, S.F. Regulation of G1 cell-cycle progression by oncogenes and tumor suppressor genes. Curr. Opin. Genet. Dev. 2002, 12, 47–52. [CrossRef] 31. Ishov, A.M.; Sotnikov, A.G.; Negorev, D.; Vladimirova, O.V.; Neff, N.; Kamitani, T.; Yeh, E.T.; Strauss, J.F., III; Maul, G.G. PML is critical for ND10 formation and recruits the PML- interacting protein daxx to this nuclear structure when modified by SUMO-1. J. Cell Biol. 1999, 147, 221–234. [CrossRef][PubMed] 32. Schneider, R.; Grosschedl, R. Dynamics and interplay of nuclear architecture, genome organization, and gene expression. Gene Dev. 2007, 21, 3027–3043. [CrossRef][PubMed] 33. Osborne, C.S.; Chakalova, L.; Brown, K.E.; Carter, D.; Horton, A.; Debrand, E.; Goyenechea, B.; Mitchell, J.A.; Lopes, S.; Reik, W.; et al. Active genes dynamically colocalize to shared sites of ongoing transcription. Nat. Genet. 2004, 36, 1065–1071. [CrossRef][PubMed] Diseases 2017, 5, 34 16 of 16

34. Peric-Hupkes, D.; Meuleman, W.; Pagie, L.; Bruggeman, S.W.; Solovei, I.; Brugman, W.; Graf, S.; Flicek, P.; Kerkhoven, R.M.; van Lohuizen, M.; et al. Molecular maps of the reorganization of genome-nuclear lamina interactions during differentiation. Mol. Cell 2010, 38, 603–613. [CrossRef][PubMed] 35. Reddy, K.L.; Zullo, J.M.; Bertolino, E.; Singh, H. Transcriptional repression mediated by repositioning of genes to the nuclear lamina. Nature 2008, 452, 243–247. [CrossRef][PubMed] 36. Kosak, S.T.; Skok, J.A.; Medina, K.L.; Riblet, R.; Le Beau, M.M.; Fisher, A.G.; Singh, H. Subnuclear compartmentalization of immunoglobulin loci during lymphocyte development. Science 2002, 296, 158–162. [CrossRef][PubMed] 37. Skok, J.A.; Brown, K.E.; Azuara, V.; Caparros, M.L.; Baxter, J.; Takacs, K.; Dillon, N.; Gray, D.; Perry, R.P.; Merkenschlager, M.; et al. Nonequivalent nuclear location of immunoglobulin alleles in B lymphocytes. Nat. Immunol. 2001, 2, 848–854. [CrossRef][PubMed] 38. Lafrenz, D.; Koretz, S.; Stratte, P.T.; Ward, R.B.; Strober, S. LPS-induced differentiation of a murine B cell leukemia (BCL1): Changes in surface and secreted IgM. J. Immunol. 1982, 129, 1329–1335. [PubMed] 39. Brooks, K.; Yuan, D.; Uhr, J.W.; Krammer, P.H.; Vitetta, E.S. Lymphokine-induced IgM secretion by clones of neoplastic B cells. Nature 1983, 302, 825–826. [CrossRef][PubMed] 40. Webb, C.F.; Das, C.; Coffman, R.L.; Tucker, P.W. Induction of immunoglobulin µ mRNA in a B cell transfectant stimulated with interleukin-5 and T-dependent antigen. J. Immunol. 1989, 143, 3934–3939. [PubMed]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).